Very pleasant! I am jealous.

Thanks haha Yes the weather is nice here.
But Thailand is very exotic, nice beaches.

 

How about just using a voltage regulator, such as an LM317HV, for your electronics.

 

HAHA nice one.
Wish I have a 43V zener, the only one with hight voltage is View attachment 240092

36V Zener plus ~4 to ~6 Volts FET Gate Threshold = ~40 to ~42 Volts.

 

36V Zener plus ~4 to ~6 Volts FET Gate Threshold = ~40 to ~42 Volts.

nd the resistance remains the same?

 

36V Zener plus ~4 to ~6 Volts FET Gate Threshold = ~40 to ~42 Volts.

Well Its working haha in the simulation when I put 41 V in the source, the simulation get crash haha but it is working when use 20 to 40 volts.
But I dont know why

 

nd the resistance remains the same?

The Resistor is just to pull the Gate-Voltage to ~zero when the Zener is not conducting.
It has only a very slight effect on the operation of the Circuit.

A FET with a "Logic-Level" Threshold-Voltage-rating will have
more Gain, and therefore a sharper-response, instead of a smoother/softer response,
but in your case a sharp-response is not really required.

If you use the FET Circuit, you must verify the Clamping-Voltage by actual measurement.
Don't just guess at it.
Look at the "Gate-Charge vs Gate-to-Source Voltage" Graph in the Data-Sheet of
which ever FET you choose, the Gate-Voltage where the line goes "flat" is
the exact "Gate-Threshold" number that you should use for your calculations.

The IRLI3705N is a 55-Volt, 52-Amp, Full-Pack-Package, Logic-Level-FET,
which has a Threshold-Voltage of only ~3.5-Volts,
and does not require an Insulator-Kit for mounting, just thermal grease.

40-Volts minus 3.5-Volts = 36.5-Volts Zener-Diode required.
39.5-Volts minus 3.5-Volts = 36-Volts Zener-Diode required.

 

The Resistor is just to pull the Gate-Voltage to ~zero when the Zener is not conducting.
It has only a very slight effect on the operation of the Circuit.

A FET with a "Logic-Level" Threshold-Voltage-rating will have
more Gain, and therefore a sharper-response, instead of a smoother/softer response,
but in your case a sharp-response is not really required.

If you use the FET Circuit, you must verify the Clamping-Voltage by actual measurement.
Don't just guess at it.
Look at the "Gate-Charge vs Gate-to-Source Voltage" Graph in the Data-Sheet of
which ever FET you choose, the Gate-Voltage where the line goes "flat" is
the exact "Gate-Threshold" number that you should use for your calculations.

The IRLI3705N is a 55-Volt, 52-Amp, Full-Pack-Package, Logic-Level-FET,
which has a Threshold-Voltage of only ~3.5-Volts,
and does not require an Insulator-Kit for mounting, just thermal grease.

40-Volts minus 3.5-Volts = 36.5-Volts Zener-Diode required.
39.5-Volts minus 3.5-Volts = 36-Volts Zener-Diode required.
.
.
.

Here in Costa Rica I found an interesting transistor.
So
36-Volts Zener-Diode + 3.8 Volts = 39.8-Volts

 

Unfortunately it's only rated for ~30-Volts, and you need at least ~50-Volts.
.
.
.

 

Unfortunately it's only rated for ~30-Volts, and you need at least ~50-Volts.
.
.
.

Oh I see. Let me see in the store.

ok, they have
this mosfet channel N with this curve Gate charge vs Gate source



So If I use the 36 V zener
the voltage break at 36 + 5 = 41V

 

That will work just fine,
But it's Threshold Voltage is 6-Volts, not 5-Volts ........

 

That will work just fine,
But it's Threshold Voltage is 6-Volts, not 5-Volts ........

Sorry mistake, thanks LowQcab

 

Today I did the first test of the PCB with the circuit and it did not work. It was quite frustrating.

I have an impedance matching problem.
When I measure the voltage that comes out of the zener regulator, it is very low compared to the value presented in simulation.

I don't know whether to use an impedance decoupling stage between the voltage regulator and the U3A amplifier.


Here is the PCB
When I measured the voltages with the voltmeter, I got 1 volt at the output of the resistor array that generates 5 volts.



Maybe you are going to tell me, that the Ian Solution its is better, but the project its advanced.

 

(25V-1V)/1k = 24mA

Looking at the graph in the LT1490 datasheet, the device needs around 35mA, and the zenar would not be getting enough current to regulate (Izt=14mA).

It's basically a case of not enough current - Try lower values for R6

 

(25V-1V)/1k = 24mA

Looking at the graph in the LT1490 datasheet, the device needs around 35mA, and the zenar would not be getting enough current to regulate (Izt=14mA).

It's basically a case of not enough current - Try lower values for R6

I am seeing that I am not reaching the minimum for a little but it is valid

 

(25V-1V)/1k = 24mA

Looking at the graph in the LT1490 datasheet, the device needs around 35mA, and the zenar would not be getting enough current to regulate (Izt=14mA).

It's basically a case of not enough current - Try lower values for R6

Nonsense - it needs 100uA - "SUpply current per amplifier" = 50uA towards the bottom of page 3.
Is the zener in back-to-front? That would give about 1V.

 

Nonsense - it needs 100uA - "SUpply current per amplifier" = 50uA towards the bottom of page 3.
Is the zener in back-to-front? That would give about 1V.

Appologies, the datasheet was approx 35uA - and that was beat case scenario temperature wise

 

Is pin 5 shorted to pin 6 and 7 on your pcb?

 

Is pin 5 shorted to pin 6 and 7 on your pcb?

I already checked and it's fine
pin 5--> Vcc
pin 6--> negative IN (Shoted with pin7)
pin 7--> output
pin 8-->postivive IN

 

Here is the PCB
When I measured the voltages with the voltmeter, I got 1 volt at the output of the resistor array that generates 5 volts.

Could you put your measured voltages on the diagram above, please.
And include the MOSFET gate and drain voltages.

 

Could you put your measured voltages on the diagram above, please.
And include the MOSFET gate and drain voltages.

Voltage gate-drain = -24.999 V
Voltage gate-source = 719.427uV